Determinants of stunting and poor linear growth in children under 2 ...

bs_bs_banner

DOI: 10.1111/mcn.12259

Original Article

Determinants of stunting and poor linear growth in children under 2 years of age in India: an in-depth analysis of Maharashtra’s comprehensive nutrition survey Víctor M. Aguayo*, Rajilakshmi Nair†, Nina Badgaiyan* and Vandana Krishna‡ United Nations Children’s Fund (UNICEF) Regional Office for South Asia, Kathmandu, Nepal, †UNICEF Field Office for Maharashtra, Mumbai, Maharashtra, India, and ‡Maharashtra Nutrition Mission, Ministry of Women and Child Development, Mumbai, Maharashtra, India *

Abstract We use a representative sample of 2561 children 0–23 months old to identify the factors most significantly associated with child stunting in the state of Maharashtra, India. We find that 22.7% of children were stunted, with one-third (7.4%) of the stunted children severely stunted. Multivariate regression analyses indicate that children born with low birthweight had a 2.5-fold higher odds of being stunted [odds ratio (OR) 2.49; 95% confidence interval (CI) 1.96–3.27]; children 6–23 months old who were not fed a minimum number of times/day had a 63% higher odds of being stunted (OR 1.63; 95% CI 1.24–2.14); and lower consumption of eggs was associated with a two-fold increased odds of stunting in children 6–23 months old (OR 2.07; 95% CI 1.19–3.61); children whose mother’s height was < 145 cm, had two-fold higher odds of being stunted (OR 2.04; 95% CI 1.46–2.81); lastly, children of households without access to improved sanitation had 88% higher odds of being severely stunted (OR 1.88; 95% CI 1.17–3.02). Attained linear growth (height-for-age z-score) was significantly lower in children from households without access to improved sanitation, children of mothers without access to electronic media, without decision making power regarding food or whose height was < 145 cm, children born with a low birthweight and children 6–23 months old who were not fed dairy products, fruits and vegetables. In Maharashtra children’s birthweight and feeding practices, women’s nutrition and status and household sanitation and poverty are the most significant predictors of stunting and poor linear growth in children under 2 years. Keywords: stunting, linear growth, children, Maharashtra, India. Correspondence: Dr Víctor M. Aguayo, United Nations Children’s Fund (UNICEF) Regional Office for South Asia, PO Box 5815, Kathmandu, Nepal. E-mail: [email protected]

Introduction Global figures indicate that 25% of children under age 5 years (i.e. 159 million) have stunted growth (United Nations Children’s Fund, UNICEF, World Health Organization, WHO, World Bank Group, WBG 2015). It is estimated that stunting – a height-for-age below minus two z-scores of the median height-for-age in the World Health Organization Child Growth Standards – is the cause of about one million child deaths annually (Black et al. 2013). For the children who survive, stunting causes lasting damage, including poor cognition and educational performance in childhood, reduced productivity and lower earnings in adulthood and, when accompanied by excessive weight gain in later childhood,

increased risk of chronic diseases (Victora et al. 2008; Dewey & Begum 2011; Black et al. 2013). India’s latest National Family Health Survey in 2006 showed that 48% of Indian children 0–59 months old were stunted (International Institute of Population Sciences (IIPS) 2007). Thus, it is estimated that at any one point an average, 61 million Indian children are stunted and therefore unable to survive, grow and develop to their full potential, which is the same potential as that of children in developed countries (Bhandari et al. 2010; World Health Organization (WHO) 2006). Recent reports indicate that the current (2014) prevalence of child stunting in India would be 38.8% (Ministry of Women and Child Development, MWCD, Government of India 2015). This means that between 2006

© 2016 The Authors. Maternal & Child Nutrition published by John Wiley & Sons Ltd Maternal & Child Nutrition (2016), 12 (Suppl. 1), pp. 121–140

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

121

122

V.M. Aguayo et al.

and 2014, the prevalence of child stunting in India declined at an average 2.4% rate annually, well above the rate of 1.7% estimated on the basis of previous surveys (International Food Policy Research Institute, IFPRI 2014). However, India remains in the category of countries with a high prevalence of stunting (30.0–39.9%) (Onis de et al. 2012). In Maharashtra – India’s second most populous state with a population over 112 million people (Office of the Registar General and Census Commissioner of India. Ministry of Home Affairs, Government of India 2011) – the poor nutrition situation of children was confirmed by India’s National Family Health Survey, which indicated that 38.8% of Maharashtra’s children 0–23 months old were stunted and over one-third of the stunted children (14.7%) were severely stunted (IIPS 2007). In response to this situation, the Government of Maharashtra created the State Nutrition Mission under the chairmanship of the State Chief Minister. The Mission was mandated to coordinate inter-sectoral efforts to reduce child undernutrition, initially (2005) in the five districts with the highest levels of child undernutrition and eventually (2009 onwards) across Maharashtra’s 35 districts. In 2012, the Government of Maharashtra commissioned an independent survey to assess progress and identify priority areas for action. The Comprehensive Nutrition Survey in Maharashtra (CNSM) showed that the prevalence of stunting in children under 2 years had declined from 38.6% in 2006 to 23.3% in 2012 (International Institute for Population Studies, IIPS 2012).

Thus, a 15.3% point decline over a 6-year period, with an average annual rate of reduction (AARR) of 2.6, significantly higher than the AARR of ~0.5 reported until 2005 (United Nations Children’s Fund, UNICEF 2013). Findings from a multidisciplinary analysis on the drivers of the decline of stunting in Maharashtra have indicated that the vision and skills of the Nutrition Mission’s leadership and staff allowed much to be accomplished, from maintaining political impetus and focus to motivating frontline workers to deliver better quality services at greater scale (Haddad et al. 2014). However, despite such significant progress, Maharashtra’s 2012 survey indicated that almost onefourth (23.3%) of children 0–23 months old were stunted and that one-third of the stunted children (7.8%) were severely stunted. Therefore, the goal of this research is to support the State Nutrition Mission to identify future policy, programme and investment priorities on maternal and child nutrition in Maharashtra through an in-depth understanding of the most important determinants of child stunting and poor linear growth. Specifically, the objective of our analysis is four-fold: (1) to characterize the epidemiology of stunting in children 0–23 months old in Maharashtra; (2) to identify the most significant predictors of stunting in children 0–23 months old; (3) to identify the most significant correlates of linear growth (height-for-age) in children 0–23 months old; and (4) to identify policy, programme and investment priorities in the context of Maharashtra’s Nutrition Mission Phase III post-2015.

Key messages • One in five (22.7%) of children 0–23 months old in the state of Maharashtra were stunted, and one-third (7.4%) of the stunted children were severely stunted. • Birthweight, child feeding, women’s nutrition and household sanitation were the most significant predictors of stunting and poor linear growth in children under 2 years. • Children born to mothers whose height was below 145 cm, had two-fold higher odds of being stunted; children born with a low birthweight had a 2.5-fold higher odds of being stunted. • Low feeding frequency and low consumption of eggs, dairy products, fruits and vegetables were associated with stunting and poor linear growth in children 6–23 months old. • Children of households without access to improved sanitation had 88% higher odds of being severely stunted.


Child stunting in Maharashtra, India

Methods We use data from the CNSM, the independent nutrition household survey conducted in 2012 at the request of the Government of Maharashtra. CNSM was designed and supervised by the International Institute for Population Studies (IIPS), the lead research agency for India’s three national Family Health Surveys – the customized version of the Demographic and Health Survey to suit the data and information needs of India – in 1992, 1999 and 2006. The representative sample of Maharashtra’s Comprehensive Nutrition Survey was designed to provide estimates of a series of key indicators on the nutrition situation of children under 2 years (0–23 months old) and their mothers in urban areas, rural areas and each of the six administrative divisions of the state: Amaravati, Aurangabad, Konkan, Nagpur, Nashik and Pune. The survey used three questionnaires: • The household questionnaire: used to collect information on all de jure (usual residents) household members, the household and the dwelling. For each person listed, information was collected on age, sex, literacy, caste/tribe and household food security and assets among other variables. • The mother’s questionnaire: administered to all women who had at least one living child in the age group 0–23 months at the time of the survey. It collected information on mother’s age, marital status, age at marriage, educational attainment, exposure to mass media, decision-making power and access to essential services among other variables. • The child’s questionnaire: administered to the mother or principal caretaker of children 0–23 months old. It was used to collect information on birth date, birthweight and feeding practices, including breastfeeding and complementary feeding practices in the 24 h preceding the survey, to assess internationally agreed Infant and Young Child Feeding (IYCF) indicators (World Health Organization, WHO, United Nations Children’s Fund, UNICEF 2008). The nutritional status of children and their mothers was assessed by measuring their height and weight following internationally agreed upon anthropometry

measurement procedures (World Health Organization, WHO 1995). A detailed description of the survey design and sample selection can be found elsewhere (IIPS 2012). In brief, a 30% prevalence of stunting in children 0–23 months old and a 10% non-response rate for anthropometry were assumed to estimate the size of the sample. The selection of the sample used a multi-stage stratified procedure. The rural sample was selected in two stages. In the first stage, villages were randomly selected with probability proportional to population size as Primary Sampling Units (PSU). In the second stage, households with at least one child 0–23 months old were randomly selected within each of the selected PSUs. In urban areas, a three-stage sampling procedure was used. In the first stage, wards were randomly selected with probability proportional to population size. In the second stage, Census Enumeration Blocks (CEB) were randomly selected with probability proportional to size. Lastly, in the third stage, a household listing was carried out in each of the selected CEB, and households with at least one child 0–23 months old were randomly selected. The survey received ethical clearance from IIPS’ Research Ethics Board. Data collection was carried out during February–April 2012. Caregivers were asked for individual consent to participate in the survey. A total of 2630 households were included in the survey. For our analysis, data from the child data set, which contains one record for every eligible child born in the 2 years prior to the survey, were used. Children with missing age and/or height were not included in the analytical sample. Stunting and severe stunting were defined as a height-for-age below 2 (moderate and severe stunting) or below 3 (severe stunting) z-scores of the median height-for-age of the World Health Organization Child Growth Standards (World Health Organization, WHO, 2006). Children with implausible height-for-age z-score (HAZ < 6 or HAZ > +6) were excluded from the analysis. In our analysis, we are interested in three outcome variables and the exposure variables that are significantly associated with them: stunting (HAZ < 2) as the indicator of choice both for surveys and global targets on child nutrition; severe stunting (HAZ < 3) to document the severity of child stunting in the population; and attained linear growth, measured as children’s HAZ.


123

124

V.M. Aguayo et al.

Analyses were performed using Stata Statistical software (College Station, TX, USA), release 12, 2011. We used sample weights to adjust standard errors for the complex survey design of CNSM. In models using stunting or severe stunting as the dependent variables, we report on odds ratios and 95% confidence intervals from logistic regression models. In models that regress the outcome variable (attained linear growth in HAZ) on exposure variables, we report on regression coefficients and 95% confidence intervals around point estimates from multiple linear regression. For all tests, p-values < 0.05 were considered statistically significant.

Findings The survey included a representative sample of 2650 children 0–23 months old. The analysis presented here pertains to 2561 children (96.6%) for whom information on age and anthropometry – and therefore on HAZ, stunting and severe stunting – was available. Children that were stunted were 22.7%, and about one-third (32.7%) of the stunted children were severely stunted (Table 1). Table 2 summarizes the socio-economic characteristics of the children included in the analysis. Households: 91.7% had access to piped water, 57.0% were food secure, 45.0% were located in rural areas, 40% were from Scheduled Castes/Scheduled Tribes and 37.9% used improved sanitation facilities. Children: a significantly higher proportion (55.2%) were boys. Most children

(91.9%) were weighed at birth, and about one in five (19.4%) of them had a low birthweight (